Leveraging social networks to fight spam

Social networks are useful for judging the trustworthiness of outsiders. An automated antispam tool exploits the properties of social networks to distinguish between unsolicited commercial e-mail - spam - and messages associated with people the user knows.     

Stronger laws needed to stem spam

Without global legislation to combat spam, the problem will never be solved despite the fact that it is costs global businesses millions of dollars a year to tackle, believes the security industry.     

融合多特征的产品垃圾评论识别

针对JINDALN等人新近提出的利用逻辑回归模型识别产品垃圾评论的检测方法中使用过多产品评论特征这一问题,分析了解决方法,并提出对特征进行显著性检验。通过对亚马逊数据集的实验结果表明,采用显著性特征建立的回归模型优于所有特征建立的模型。新模型不仅解决了上述问题,减少了计算量,而且整体性能不变,这表明以显著性特征建模有助于提高模型的检测质量。     

Using online linear classifiers to filter spam emails

The performance of two online linear classifiers—the Perceptron and Littlestone’s Winnow—is explored for two anti-spam filtering benchmark corpora—PU1 and Ling-Spam. We study the performance for varying numbers of features, along with three different feature selection methods: information gain (IG), document frequency (DF) and odds ratio. The size of the training set and the number of training iterations are also investigated for both classifiers. The experimental results show that both the Perceptron and Winnow perform much better when using IG or DF than using odds ratio. It is further demonstrated that when using IG or DF, the classifiers are insensitive to the number of features and the number of training iterations, and not greatly sensitive to the size of training set. Winnow is shown to slightly outperform the Perceptron. It is also demonstrated that both of these online classifiers perform much better than a standard Naïve Bayes method. The theoretical and implementation computational complexity of these two classifiers are very low, and they are very easily adaptively updated. They outperform most of the published results, while being significantly easier to train and adapt. The analysis and promising experimental results indicate that the Perceptron and Winnow are two very competitive classifiers for anti-spam filtering.     

Behavior-based approach to detect spam over IP telephony attacks

Spam over IP telephony (SPIT) is expected to become a serious problem as the use of voice over IP grows. This kind of spam is appreciated by spammers due to its effectiveness and low cost. Many anti-SPIT solutions are applied to resolve this problem but there are still limited in some cases. Thus, in this paper, we propose a system to detect SPIT attacks through behavior-based approach. Our framework operates in three steps: (1) collecting significant calls attributes by exploring and analyzing network traces using OPNET environment; (2) applying sliding windows strategy to properly maintain the callers profiles; and (3) classifying caller (i.e., legitimate or SPITter) using ten supervised learning methods: NaïveBayes, BayesNet, SMO RBFKernel, SMO PolyKernel, MultiLayerPerceptron with two and three layers, NBTree, J48, Bagging and AdaBoostM1. The results of our experiments demonstrate the great performance of these methods. Our study, based on receiver operating characteristics curves, shows that the AdaBoostM1 classifier is more efficient than the other methods and achieve an almost perfect detection rate with acceptable training time.     

Distrust seed set propagation algorithm to detect web spam

Web spam uses numerous techniques to misguide Web search engines in exchange of financial profit. A myriad of semi-automatic propagation model has been proposed with the purpose of combating Web spam. In this paper, distrust propagation is used to detect Web spam. An automatic distrust seed set propagation algorithm (DSP), which acts as an extension to the seed set to propagate distrust further to detect more Web spam. Experiments are conducted on WEBSPAM-UK2006 and WEBSPAM-UK2007 dataset; the results have shown that DSP enhanced the baseline algorithms and detected 17.73 % more spam hosts in the former dataset and detected 8.59 % more spam hosts in later dataset.     

Application of morphological filters to tackle discretisation artefacts

A post-processing technique for removing discretisation artefacts (jaggies) is described. It is based on the detection of a jaggy by means of a suitable morphological filter and then using a heuristic algorithm to compute a probable density distribution, which is applied to the pixels adjacent to the jaggy. Also dashed lines are recognized and interpreted as possible discretisation artefacts. The method applies to both spatial and temporal aliasing.     

降级Web Spam的可信度链接分析算法

以降级Web Spam算法为研究内容,在分析TrustRank等算法的优点和不足的基础上,提出了时间可信度的概念刻画不同时间下页面的可信度,引入了CreditRank算法来计算页面的可信度.同时,引入了LinkRank算法来计算基于链接的页面质量,整合权威度和可信度.定性分析和实验结果表明,CreditRank算法扩展了种子的利用范围,提高了反Spam算法的覆盖度,LinkRank算法能够解决"无辜页面"问题,并取得很好的降级Spam的效果.     

Techniques to tackle state explosion in global predicate detection

Global predicate detection, which is an important problem in testing and debugging distributed programs, is very hard due to the combinatorial explosion of the global state space. The paper presents several techniques to tackle the state explosion problem in detecting whether an arbitrary predicate Φ is true at some consistent global state of a distributed system. We present space efficient online algorithms for detecting Φ. We then improve the performance of our algorithms, both in space and time, by increasing the granularity of the execution step from an event to a sequence of events in each process